Project Summary/Abstract. The long-term objectives of this proposal are to define genotype/phenotype correlations for the extended Prader-Willi/Angelman domain on chromosome 15q11-q13 that stretches from Breakpoint 1 (BP1) to Breakpoint 5 (BP5), with special emphasis on the BP4-BP5 region where deletions are associated with mental retardation, autism, epilepsy, schizophrenia, and bipolar disorder. Genetic analysis will focus both on copy number variations (CNVs) deleting or duplicating multiple contiguous genes and on point mutations within individual genes, especially CHRNA7. Previous work on this project has focused on Prader- Willi syndrome (PWS) and Angelman syndrome (AS), caused by paternal and maternal deficiency, respectively, for the central portion of the domain. The focus is now being broadened to include autism caused by duplications of this region and to study phenotypes associated with the flanking portions of the domain, BP1 to BP2 and BP4 to BP5. Deletions of chromosome 15q13.3 (BP4-BP5) remove six contiguous genes, and a smaller deletion removes CHRNA7 and one exon of an adjacent gene. Aim 1 is to determine genotype/phenotype relationships for the BP4-BP5 region with emphasis on the CHRNA7 gene. Aim 1a is to determine if the 15q13.3 and CHRNA7 duplications are pathological or benign using case control studies. Aim 1b is to identify the basis for the phenotypic heterogeneity associated with the 15q13.3 and CHRNA7 deletions focusing primarily but not exclusively on various genetic modifier effects. Aim 1c is to search for loss-of- function point mutations in CHRNA7 in particularly high risk samples such NIMH samples with both epilepsy and schizophrenia or both epilepsy and bipolar disorder. Aim 2 is to determine genotype/phenotype relationships for the BP1-BP2 region and the four genes therein. Aim 3 is to determine the molecular basis for the parent of origin effects of duplications of the BP1/BP2 to BP3 region which typically cause autism when on a maternal chromosome and are usually benign when on the paternal chromosome. This aim will emphasize expression analysis (increasingly relying on RNA-Seq) and epigenetic studies of human brain tissue comparing duplications of the 15q11-q13 region with controls. Aim 4 is to determine the function of the snoRNA HBII-85 cluster, because there is now relatively strong evidence that paternal deficiency for this snoRNA cluster causes the major components of the PWS phenotype. This aim will focus on genome-wide analysis of expression and on alternative splicing and RNA editing (again using RNA-Seq) in human and mouse brain lacking expression of HBII-85. A series of target candidate genes have been identified by others based on bioinformatic analysis, and further bioinformatic analysis is planned. The expression of these candidate genes will be studied in control and PWS brain from mouse and human using RT-PCR to analyze alternative splicing and RNA editing.